Pelotas Birth Cohort Study

South America

Perinatal study All children born in the maternity hospitals Family living in the urban area of the city Hospital deliveries accounted for more than 99% of births Losses < 1%

Questionnaire

1983 Follow-up study All children born from Januay to April 1982 (1/3 of the cohort) Visit to adressess obtained in hospital Losses = 21% Age = 11.3 months

1984 Follow-up study All cohort children Census of all households in the city (70,000) Losses = 13% Age = 19.4 months

1986 Follow-up study Age = 43.1 months Losses = 16% All cohort children Census of all households in the city (70,000) Age = 43.1 months Losses = 16%

1997 Follow-up visit Visit to all households in a systematic sample of 27% of the city’s census tracts (70) Losses = 28.2% Age = 14.7 years

2000 Follow-up visit All male cohort adolescents Losses = 21% Army recruitment examination Losses = 21% Age = 18.2 years

2001 Follow-up visit Losses = 31% Age= 18.9 years Visit to all households in a systematic sample of 27% of the city’s census tracts (70)

2004-5 Follow-up visit Census of all households in the city (98,000) plus visit to previous addresses Losses: 22.6% Age = 23 years

20012-13 Follow-up visit Subjects were searched using multiplie strategies Invited to visit the research clinic Losses: 31.9% Age = 30 years

Questionnaire – 2012/13 Eletronic questionnaire 587 questions Physical examination Blood sample

What has been measured? Early phases Socioeconomic and demographic variables Maternal characteristics Pregnancy Delivery Mortality Environmental characteristics Health service utilization and morbidity Nutrition Child care Maternal health Psychosocial development

What has been measured? Late phases Family situation Schooling Dietary habits Smoking Physical activity Illicit drug use Use of health services Offspring Weight Standing and sitting height Subscapular and triceps skinfold Body composition (bioimpedance) Waist circumference Blood pressure Blood lipids C reactive protein

Measurements – (2012) Body composition – 4 compartments model Intima media thickness Visceral fat Bone mineral density - DXA Peak flow velocity IQ Depression

Challenges and strategies

1983 visit Visit to addresses obtained in hospital Losses: 21 % Address did not exist Family was not known in the address Moved We had gathered information on the mother’s name and family address

What we did We tried to find any information on the family address Neighbours Electric Power Company Department store

1984 New follow-up strategies Census in the urban area 70,000 households were visited In search of children born in 1982 Linkage of datasets Perinatal study Census

Fuzzy logic - Manual Hospital Date of birth Maternal name

Losses to follow-up

490 losses in 1984 and 1986 5914 livebirths 233 deaths 266 - 1986 449 - 1984 4476 evaluated in 1984 e 1986

Losses – The return 2004 Census - identified only 52% of the cohort members

Other strategies Visit to the last address SINASC – National Birth Registration Service Army recruitment Office University and Middle Schools Electoral register

Challenge Lack of Funding Precluded regular visits Followed a sampling fraction of the cohort 930 subjects were followed in 1984, 1986, 1997 e 2001.

Visits – 4 cohorts ● 1,3, 6 and 12 months ● 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 1982 ● 1993 2004 2015 ● 1,3, 6 and 12 months 3 and 12 months ● Sample